I was pretty stoked about getting the chance to interview Graywolf about the Google landing page quality update. There has been a lot said about the latest update and I wanted to hear the thoughts of someone who has really been impacted by the change. Graywolf has been one of the biggest critics of the landing page quality score and he’s also been the most open about sharing his own data in the public domain.
I posed some questions to Michael below which are as you might expect if you have read his blog or posts elsewhere before; open, to the point and extremely enlightning. Thanks again to Michael for
answering these questions.
What do you find frustrating about the latest Google landing page quality update?
I have to say the most frustrating part of the entire change has been the lack of disclosure of what they are really looking for and saying is high quality. As an advertiser you measure your success by ROI, how much did spend, how much did I make, how much profit was there and was it worth the time and resources you invested. Google is measuring multiple factors and being vague about what they are and what role they play. From where I stand it would be much easier for everyone if they just said we want X, Y, and Z, I’m sure 99% of the people would just make the changes they wanted.
It seems that many in the industry believe that the latest landing page quality update is not a real attempt by Google in improving quality of user experience by limiting MFA’s, arbitragers and other affiliates. What are your thoughts?
IMHO the changes aren’t targeted at all affiliates, just “thin affiliates”. For example Google doesn’t see any real value in having 6 Coca-Cola vending machines right next to each other. As an advertiser if I can put a flashier front panel on my vending machine enticing you chose it over the other 5 I’m willing to take that risk. I’m sure some less sophisticated mom & pop businesses who fit loosely into the “affiliate fingerprint” got caught in the trap which is really unfortunate.
Arbitragers seem to be a particular focus for Google right now. If Google wanted to eradicate arbitragers, could it not do so by enforcement of its own advertiser/publisher policies, terms & conditions?
There’s a few different kinds of arbitrage, most people understand the contextual arbitrage model, however there’s also the acquisition and lead generation that I classify as arbitrage as well. If you’re paying for the traffic and not ultimately completing the conversion, you come under the arbitrage model. The problem is lots of small business use hosted carts, pay pal, 2checkout or other off site services, so it’s hard to create public rule set that doesn’t have more exceptions and holes than a block of Swiss cheese.
Some have suggested that Google is being far to authoritarian in appointing themselves as the rightful judge for everyone in what they may consider ‘quality’. How do you believe Google is determining this ‘quality’?
The use of the word ‘quality’ was probably not the best choice in hindsight, as everyone has a different set of standards. For example I may consider a Honda or Toyota a quality car, but Donald Trump has a different set of requirements for quality, it’s all very subjective. Conversely you can’t come out and say we’re going to judge your ads by our arbitrary set of criteria, and not expect to look foolish. So using the word ‘quality’ was window dressing that allowed them some editorial latitude and freedom behind the curtain.
Are they too authoritarian? Well I do think Google is still a bit arrogant. They are a smart bunch of folks and they realize it’s there’s very little personal cost involved in switching search engines. That’s why they are developing other services to blur the lines between Google and the Internet, like Gmail, notebooks, writely, spreadsheets and any of the other services. The cost (non monetary) of moving all of those services and data is much larger and you are much less likely to make the change impulsively or on a whim.
What ppc campaigns have you seen worst hit by the latest update? Are they the same campaigns hit by the original update?
The sectors that I suffered the most damage in are the ones that are chock full of affiliates, some were the same as those hit in the first change some weren’t. To a certain extent it’s a sector based, but I just think they are fine tuning their definition of ‘quality’.
Time to dazzle us with some stats. How many campaigns and keywords do you work with? Do you know a rough percentage of your keywords that have been impacted?
The first quality change hit me really hard I lost probably about 90% of my keywords in one day. Since I got into PPC to stabilize my income I was understandably pretty upset. I hunkered down drank lots of red bull, spent some long late nights testing different things until I found a way to get things to work. It took me about 3 weeks to get back the biggest chunk of those keywords, but some things I had been bidding on for years I never was able to get back. When I made those changes I fully expected another set of changes in the coming months and set things up with that in mind. So when they made the change in November I lost about 60% of my words, however I was less upset as it took me only two days of grunt work to get almost everything working again.
I fully expect there to be other quality adjustments in the future, and I think I’ve taken steps to be prepared. The thing to get out of all that is to evolve or die. With each iteration I become smarter more efficient and more agile. As far as how many keywords, I test at least one two new programs a month, so work some don’t, I’m probably bidding on about 20,000 terms right now.
Is it mainly pages with contextual advertising that you have seen hit by the latest update?
If this change was just about contextual arbitrage it would be pretty easy to profile, from my testing they are looking at other factors.
You have publicised some interesting stats over at Threadwatch showing keywords with high CTR and conversion being hit by the latest update. Google has recently issued a firm denial that it uses CTR or conversion in calculating minimum bids. Do you believe this to be the case?
I know a lot people don’t like it or don’t want to hear it but I believe they are incorporating user data and parts of the organic algorithm into the quality score. It’s much more than just high CTR and or conversion rates that keep you from getting price jacked. Since the organic algo is updated and tweaked every few months if they were using it we should also see updates to the quality scores at periodic intervals during the next year, so we’ll have to stay tuned and watch.
Have you tested tracking your conversion outside of Google?
Hypothetically speaking I don’t think the really smart people are telling Google everything, instead I think they are giving Google the data they think Google wants to see, hypothetically speaking of course. Additionally I also think those smart people aren’t giving that level of data to the merchants. Which is why I’m not a big fan of overly complex merchant reporting systems
Have you tested your landing pages with a view to increasing quality and reducing min cpc?. Are there any particular interesting findings that you can share?
I’ve never seen any changes in pricing based on changing minor elements on the landing page. Increasing your density, bolding text, putting it in an H1 tag, just not really not worth the time fiddling with.
This question was posed to Google recently; Can a page that has a high CTR or conversion rate be considered a ‘poor’ quality landing page?
Going back to my vending machine analogy, Google wants to build a community with store keepers who are in it for the long haul, and are willing to put down roots and are concerned about the long term viability of the community. Affiliates just want a traffic and are content to put a vending machine it up wherever there are a lot of people. For Google these changes are about their long term survival as a quality website, as an affiliate to be honest you don’t care if the traffic is from Google, Yahoo, MSN, ASK or anywhere else, as long as you can get it, it converts and you make a profit. So put some bright lights and an eye catching panel on your vending machine people will come over to it (high CTR) as long as the soda is the flavor the customer wants, is not sold out and cold they will buy it (high conversion) but there’s very little “user experience” involved in that transaction.
Having the landing page quality algo completely independent to the rest of the Quality Score that makes up ad rank provides Google completely separate control over their minimum bids. How do you feel this will continue to effect advertisers & shape the industry moving forward?
It’s actually quite bizarre and frankly has some ridiculous outcomes. For example let’s say you get labeled as low quality and your bid gets raised to $5 or $10. Let’s say you are kind of crazy and decide to pay $5 or $10 what happens is you jump up to the #1 slot since everyone else is bidding $0.30 or $0.40 a keyword. Now the kicker is you get charged the full $5 or $10 not just a bit more than the bidder below you. The whole “your ad is low quality unless you pay through the nose” is actually ridiculous, and makes them look like they are in it for the money. If my ad is low quality don’t show it, plain and simple, end of story.
Is there anything else you would like to add or share with us?
People who were priced out of their keywords need to try things differently. Don’t be afraid to try something that’s out of the box and little crazy, just watch it like a hawk when you do and see what happens. If you can’t figure out what’s working try looking at people who have campaigns that didn’t get disabled and try to dissect it.
Thanks again to Graywolf!